The use of multiple material fluorescence-based sensors, where each is optimized to a particular temperature range yet is pumped by the same light source, emitting over the same spectral region, makes for a very simple, convenient and promising optical arrangement which can be applied in real-time, quasidistributed temperature sensor systems. The fluorescence lifetime approach, which is an important technique to enable fluorescence emission to be exploited for thermometry, is adopted in the system discussed. An analysis scheme using Prony’s method has been reported which enables exponential decays from either single-material or two material and quasidistributed sensors to be deconvolved and thus data and associated measurand information encoded in each individual signal to be recovered. In this work, in the development of quasidistributed temperature sensor algorithms based on Prony’s method are used for the estimation of exponential time constants of a convolved triple exponential fluorescence decay, each corresponding to a different-point temperature. Experimental results obtained are presented to justify their use in practical multiexponential fluorescence decay analysis and show a comparison of the Prony method to the Marquardt nonlinear least-squares approximation algorithm to achieve the deconvolution. The computational time for Prony’s approach is approximately one-thousandth that of the Marquardt technique while the accuracy achieved using Prony’s method is still high enough for practical use. ©1998 American Institute of Physics.